Apositive family history is associated with an increased. Helicobacter pylori Infection Among Offspring of Patients With Stomach Cancer

Transcription

1 GASTROENTEROLOGY 2000;118:31 35 Helicobacter pylori Infection Among Offspring of Patients With Stomach Cancer HERMANN BRENNER,* GÜNTER BODE, and HEINER BOEING Departments of *Epidemiology and Internal Medicine I, University of Ulm, Ulm; and Department of Epidemiology, German Institute of Human Nutrition, Potsdam-Rehbrücke, Germany See editorial on page 222. Background & Aims: A positive family history is associated with an increased risk of stomach. We compared the prevalence of Helicobacter pylori infection, a known risk factor for stomach, between subjects with and without parental history of stomach to evaluate a potential role of H. pylori infection in familial aggregation of stomach. Methods: A total of 1351 men and women aged years who participated in the German Health and Nutrition Survey conducted in the western part of Germany in were included in the study. Detailed information on sociodemographic factors, nutritional factors, and parental history of was obtained by standardized interviews. Serum samples were analyzed for immunoglobulin G antibodies against H. pylori by enzyme-linked immunosorbent assay. Results: The prevalence of H. pylori infection was much higher (69%) among subjects with a parental history of stomach than among other subjects (44%). This association persisted after control for potential confounders by multiple logistic regression (adjusted odds ratio, 2.7; 95% confidence interval, ), and was particularly strong among subjects below age 55 (adjusted odds ratio, 5.1; 95% confidence interval, ). Conclusions: These results suggest that familial aggregation of stomach may be explained at least partly by familial clustering of H. pylori infection. Apositive family history is associated with an increased risk of stomach. 1 4 Familial aggregation of stomach may reflect common genetic and common environmental risk factors among patients with stomach and their family members. An important environmental risk factor for stomach is infection with the gastric spiral bacterium Helicobacter pylori. In a recent meta-analysis of 19 cohort and case-control studies, a summary odds ratio (OR) of 1.92 has been found, suggesting an approximately 2-fold risk of stomach among infected individuals. 5 This meta-analysis also showed that the relationship between H. pylori infection and stomach varies strongly by age, showing a much stronger relationship among younger than among older subjects. It is also suggested that stomach is more strongly linked with H. pylori strains that have the cytotoxin-associated gene (caga), which encodes a highly immunogenic high-molecularweight (120 kilodaltons) protein (CagA), than with other H. pylori strains. Several studies have shown that H. pylori infection tends to cluster within families. In particular, there is increasing evidence of a major role of parent-child transmission of the infection We assessed the prevalence of H. pylori infection among offspring of parents with stomach, offspring of parents with other forms of, and offspring of parents who did not develop to evaluate a potential role of H. pylori infection in familial aggregation of stomach. Materials and Methods Study Design Our analysis is based on a substudy of the German National Health and Nutrition Survey that was carried out in the western part of Germany in (VERA-Study, Verbundstudie Ernährungserhebung und Risikofaktoren Analytik). Details of the study design have been reported previously. 13 Briefly, a 3-stage (electorial districts, households, individuals) stratified probability sample was drawn from the noninstitutionalized adult population ( 18 years) of German nationality. With a participation rate of approximately 70%, 2006 men and women were recruited in the study. Data Collection Personal interviews were conducted at participants homes by trained personnel. Subjects were interviewed about their own medical history, sociodemographic factors, and dietary and lifestyle habits. In addition, participants were asked if their fathers or mothers ever had and, if so, what type of. Abbreviations used in this paper: caga, cytotoxin-associated gene; 95% CI, 95% confidence interval; OR, odds ratio by the American Gastroenterological Association /00/$10.00

2 32 BRENNER ET AL. GASTROENTEROLOGY Vol. 118, No. 1 Blood samples were obtained by venipuncture on the morning of examination after fasting overnight from 1834 subjects. Serum samples were stored in the gas phase of liquid nitrogen at 130 C and analyzed for H. pylori immunoglobulin G antibodies by enzyme-linked immunosorbent assay (GAP; Bio-Rad Laboratories Diagnostics Group, München, Germany) in a central laboratory at the University of Ulm in According to the manufacturer s instructions, levels 20 U/mL were considered positive and levels 12.5 U/mL were considered negative. Borderline levels between 12.5 and 20 U/mL were treated as negative in the statistical analyses (alternative analyses excluding these individuals yielded very similar results; therefore, they are not reported separately). The GAP Bio-Rad immunoglobulin (Ig) G test kit has been evaluated extensively, and it performed well compared with other commercial kits. Median sensitivity and specificity reported by 1997 among 20 populations including a total of 2877 subjects were 94% and 79%, respectively. 14 The serum of IgG-positive individuals was further analyzed for the presence of antibodies against the CagA antigene using a commercial Western blot (Helicobacter pylori Westernblot; AID, Strasberg, Germany; distributed by DPC Biermann, Bad Nauheim, Germany) as previously described. 15 The study was approved by the ethics committee of the University of Ulm. Statistical Analysis The current analysis was restricted a priori to subjects aged years (n 1351). None of the subjects below age 30 (n 453) had a family history of stomach. Subjects aged 75 or older (n 30) were excluded to reduce potential bias by poor specificity of serological ascertainment of H. pylori infection 16 as well as potentially less accurate ascertainment of parental stomach in this age group. We first described subjects with a parental history of stomach (i.e., subjects who reported that either their father or their mother had stomach ), subjects with a parental history of other s, and subjects with no parental history of according to basic sociodemographic factors. We then compared the prevalence of H. pylori infection in the 3 groups of study participants. We used multiple logistic regression to assess the relationship between parental history of stomach or other and H. pylori infection (dependent variable) while controlling for potential confounding factors. The following covariates, which are known or suspected to be related to H. pylori infection, were considered 17,18 : age (in years), gender, years of education (categories: 9 years, years, 12 years; these categories reflect standard categories of the German school system, and they are closely related to both childhood and adulthood socioeconomic status in Germany), and alcohol consumption (in grams per day; derived from average consumption of various alcoholic beverages according to food frequency questionnaire). In addition to assessment of the overall relationship between parental history of and H. pylori infection, separate analyses were carried out for maternal and paternal history of. Furthermore, we performed separate analyses in younger (30 54 years) and older (55 74 years) subjects because the role of both family history of and H. pylori infection is known to vary strongly by age. Finally, the relationship between parental history of and H. pylori infection was analyzed separately for CagA-positive and CagA-negative H. pylori serology. All analyses were carried out on a personal computer with the SAS statistical software package. Results Overall, 39 of 1351 subjects (2.9%) reported a history of stomach among either of their parents, and 291 (21.5%) subjects reported a parental history of another type of. As shown in Table 1, median age was 48 years, and 56% of subjects were female. Subjects with a parental history of stomach were somewhat older (median, 56 years) and more often female (69%) than subjects with no parental history of (median age, 47 years; 54% female) or with a parental history of another (median age, 50 years; 61% female). School education and median alcohol consumption were similar in all 3 groups. Overall, 608 subjects (45%) had a positive IgG response to H. pylori infection. Table 2 shows the prevalence of H. pylori infection according to parental history of in the entire sample. The prevalence of H. pylori infection was much higher in subjects with a parental history of stomach (69%) than in those with no parental history of. Compared with the latter group, the crude odds ratio (OR) of H. pylori infection among subjects with a parental history of stomach was 2.9 (95% confidence interval [95% CI], ). This association was only slightly reduced Table 1. Parental History of Cancer and Basic Characteristics of Study Population History of among parents Stomach (n 39) Other (n 291) No (n 1021) Total (n 1351) Age ( yr) Median Interquartile range Sex % Female 69% 61% 54% 56% School education 9 yr 59% 56% 56% 56% yr 26% 28% 28% 28% 12 yr 15% 16% 16% 16% Alcohol consumption ( g/day) Median Interquartile range

3 January 2000 PARENTAL STOMACH CANCER AND H. PYLORI INFECTION 33 Table 2. H. pylori Infection by History of Cancer Among Parents HP OR (95% CI) History of among parents n n % Crude Adjusted a Father or mother No % 1.0 ref 1.0 ref % 1.1 ( ) 1.0 ( ) Stomach % 2.9 ( ) 2.7 ( ) Mother No % 1.0 ref 1.0 ref % 1.2 ( ) 1.1 ( ) Stomach % 4.8 ( ) 3.2 ( ) Father No % 1.0 ref 1.0 ref % 1.0 ( ) 0.9 ( ) Stomach % 1.8 ( ) 2.2 ( ) HP, H. pylori positive; ref, reference group. (OR, 2.7; 95% CI, ) after multiple logistic regression analysis was performed to control for potential confounders. No association was seen between a parental history of nonstomach and prevalence of H. pylori infection. The association with H. pylori infection was slightly more pronounced for maternal history of stomach than for paternal history of stomach, but these differences must be interpreted with caution because of the small numbers of subjects with a parental history of stomach and the wide confidence intervals in the parent-specific analyses. Among the majority of patients with no parental history of, prevalence of H. pylori infection was much lower in subjects aged (36%) than in those aged (59%) (Table 3). In contrast, the prevalence of infection was high in both age groups (and somewhat higher still in the younger age group) among subjects with a parental history of stomach. Therefore, the association of a parental history of stomach with H. pylori infection was much stronger in the younger group (adjusted OR, 5.1; 95% CI, ) than in the older group, in which it was not statistically significant (adjusted OR, 1.2; 95% CI, ). As shown in Table 4, a parental history of stomach was positively related to both CagA-negative and CagA-positive H. pylori serology, although the relationship with CagA-positive serology was somewhat weaker and not statistically significant given the small numbers of subjects in the subgroup-specific analyses. Discussion In this large population-based study, offspring of patients with stomach had a clearly increased prevalence of H. pylori infection as determined by an extensively evaluated commercial IgG enzyme-linked immunosorbent assay. 14 The association of a parental history of with H. pylori infection among the offspring was specific for stomach, and it was particularly pronounced among younger study participants. These results support the hypothesis that familial aggregation of stomach 1 4 may be at least partly mediated by intrafamilial clustering of H. pylori infection The association of stomach with H. pylori infection in the offspring observed in this study is of similar magnitude or even stronger than the association observed between stomach and subjects own H. pylori infection in most of the epidemiological studies. 5 Although the correlation of H. pylori infection between parents and their offspring tends to be very high, 6 12 it is certainly less than perfect. Therefore, stomach would be expected to be more strongly related to patients own H. pylori infection than to H. pylori infection in the offspring. However, the former association may be underestimated more often because patients with stomach generally have more severe mucosal atrophy and intestinal metaplasia in the stomach than Table 3. H. pylori Infection by Age and History of Cancer Among Parents HP OR (95% CI) Age ( yr) History of in father or mother n n % Crude Adjusted a No % 1.0 ref 1.0 ref % 1.2 ( ) 1.1 ( ) Stomach % 5.0 ( ) 5.1 ( ) No % 1.0 ref 1.0 ref % 0.8 ( ) 0.9 ( ) Stomach % 1.3 ( ) 1.2 ( ) HP, H. pylori positive; ref, reference group.

4 34 BRENNER ET AL. GASTROENTEROLOGY Vol. 118, No. 1 Table 4. CagA-Negative and -Positive H. pylori Serology by History of Cancer Among Parents History of in father or mother H. pylori infection OR (95% CI) a No CagA CagA CagA CagA No 575 (56%) 178 (17%) 268 (26%) 1.0 ref 1.0 ref 156 (54%) 59 (20%) 76 (26%) 1.1 ( ) 1.0 ( ) Stomach 12 (31%) 15 (38%) 12 (31%) 3.8 ( ) 1.9 ( ) ref, reference group. healthy subjects, 19,20 conditions that increase the chance of underdetection or spontaneous loss of infection. 21,22 Another intriguing question is to what extent familial aggregation of stomach may be explained by familial clustering of H. pylori infection. With a relative risk of stomach of approximately 2 for infected compared with uninfected subjects, the prevalence of H. pylori infection in subjects with and without parental history of stomach observed in our study (69% and 44%, respectively) would translate to a 17% increase [( )/( )] in the expected occurrence of stomach in the former over the latter group. Even if the true relative risk associated with H. pylori infection was as large as 5, the overall occurrence of stomach in subjects with a positive family history would be expected to be only 36% higher than in subjects with no family history. These figures are lower than the relative risks of stomach of that have been found for a positive family history of this disease in pertinent epidemiological studies. 1 4 Although comparability between studies is limited given differences in methods of determining parental history, changing types of gastric (proximal vs. distal) over time, and differences in overall prevalence of H. pylori infection between populations, this pattern suggests that other factors, such as nutritional and genetic factors, which also tend to cluster within families, are likely to be important as well. In our study, a family history of stomach was associated with an overall increase in the risk of H. pylori infection, but this association was not stronger for CagA-positive than for CagA-negative H. pylori serology as evaluated by a widely used commercial Western blot. 15 If anything, the reverse was true. This finding appears to conflict with those of several recent studies that have reported a higher risk of stomach associated with CagA-positive H. pylori infections However, results have not been consistent, 30 and there are suggestions that CagA is not directly involved in the pathogenesis of stomach. 31 A limitation of our study that is shared with most other epidemiological studies is that H. pylori infection status could only be determined serologically. Although the sensitivity of the commercial test used in our study was found to be very high, 14 its limited specificity may have led to some underestimation of the relationship between parental history of stomach and H. pylori infection. 32 Another limitation of our study was that the exposure variable of primary interest, i.e., family history of stomach, was based on self-reports without a possibility of validation by medical records. Potential misclassification of family history may likewise have led to some underestimation of the relationship with H. pylori infection. 32 This may have applied particularly to older subjects, which might at least partly explain the much lower association observed with H. pylori infection in this group. Interestingly, the variation of results between age groups in our study closely parallels agerelated patterns found in studies on the relationship between stomach and the subjects own H. pylori infection status. Lower accuracy of serological testing for H. pylori infection and a higher rate of spontaneous loss of the infection in older than in younger subjects may contribute further to these patterns. 16 Another limitation of our study is the small number of subjects with a family history of stomach, leading to rather wide confidence intervals for the estimates of association. Despite the small numbers, however, the associations were clearly statistically significant, making chance a very unlikely explanation for the observed patterns. Therefore, despite its limitations, our study shows a strong relationship between parental stomach and H. pylori in the offspring. This finding supports a role of H. pylori infection beside other factors in familial aggregation of stomach, and it may have important clinical implications. Subjects with both a family history of stomach and H. pylori infection may be at particularly high risk of stomach. Because H. pylori infection, unlike other causes of familial aggregation of stomach, may be treated effectively, our results suggest that screening for and eventual treatment of H. pylori infection may be a promising approach to reduction of the increased risk of stomach in offspring of patients with this disease. However, further research, particularly controlled clinical trials, is needed to allow

5 January 2000 PARENTAL STOMACH CANCER AND H. PYLORI INFECTION 35 for comprehensive evaluation of potential benefits of pertinent screening strategies. 33 References 1. Graham S, Lilienfeld AM. Genetic studies of gastric in humans: an appraisal. Cancer 1958;11: La Vecchia C, Negri E, Franceschi S, Gentile A. Family history and the risk of stomach and colorectal. Cancer 1992;70: Palli D, Galli M, Caporaso NE, Cipriani F, Decarli A, Saieva C, Fraumeni Jr JF, Buiatti E. Family history and risk of stomach in Italy. Cancer Epidemiol Biomarkers Prev 1994;3: Inoue M, Tajima K, Yamamura Y, Hamajima N, Hirose K, Kodera Y, Kito T, Tominaga S. Family history and subsite of gastric : data from a case-referent study in Japan. Int J Cancer 1998;76: Huang J-Q, Sridhar S, Chen Y, Hunt RH. Meta-analysis of the relationship between Helicobacter pylori seropositivity and gastric. Gastroenterology 1998;114: Mitchell HM, Bohane TD, Berkowitz J, Hazell SL, Lee A. Antibody to Campylobacter pylor in families of index children with gastrointestinal illness due to Campylobacter pylori. Lancet 1987;2: Malaty HM, Graham DY, Klein PD, Evans DG, Adam E, Evans DJ. Transmission of Helicobacter pylori infection. Studies in families of healthy individuals. Scand J Gastroenterol 1991;26: Drumm B, Pérez-Pérez GI, Blaser MJ, Sherman PM. Intrafamilial clustering of Helicobacter pylor infection. N Engl J Med 1990;322: Bamford KB, Bickley J, Collins JSA, Johnston BT, Potts S, Boston V, Owen RJ, Sloan JM. Helicobacter pylori: comparison of DNA fingerprints provides evidence for intrafamilial infection. Gut 1993;34: Wang J-T, Sheu J-C, Lin J-T, Wang T-H, Wu M-S. Direct amplification and restriction pattern analysis of Helicobacter pylori in patients with duodenal ulcer and their families. J Infect Dis 1993;168: Blecker U, Lanciers S, Mehta DI, Vandenplas Y. Familial clustering of Helicobacter pylor infection. Clin Pediatr 1994;33: Rothenbacher D, Bode G, Berg G, Knayer U, Gonser T, Adler G, Brenner H. Helicobacter pylori among preschool children and their parents: evidence of parent-child transmission. J Infect Dis 1999;179: Heseker H, Hartmann S, Kübler W, Schneider R. An epidemiologic study of food consumption habits in Germany. Metabolism 1995;44(suppl 2): Laheij RJF, Straatman H, Jansen JBMJ, Verbeek ALM. Evaluation of commercially available Helicobacter pylori serology kits: a review. J Clin Microbiol 1998;36: Rokkas T, Ladas S, Liatsos C, Petridou E, Papatheodorou G, Theocharis S, Karameris A, Raptis S. Relationship of Helicobacter pylori CagA status to gastric cell proliferation and apoptosis. Dig Dis Sci 1999;44: Liston R, Pitt MA, Banerjee AK. IgG ELISA antibodies and detection of Helicobacter pylori in elderly patients. Lancet 1996; 347: Rosenstock SJ, Andersen LP, Rosenstock CV, Bonnevie O, Jørgensen T. Socioeconomic factors in Helicobacter pylori infection among Danish adults. Am J Public Health 1996;86: Brenner H, Berg G, Lappus N, Kliebsch U, Bode G, Boeing H. Alcohol consumption and Helicobacter pylori infection: results from the German National Health and Nutrition Survey. Epidemiology 1999;10: Sipponen P, Kekki M, Haapakoski J, Ihamäki T, Siurala M. Gastric risk in chronic atrophic gastritis: statistical calculations of cross-sectional data. Int J Cancer 1985;35: Kato S, Onda M, Matsukura N, Tokunaga A, Matsuda N, Yamashita K, Shields P. Genetic polymorphisms of the related gene and Helicobacter pylor infection in Japanese gastric patients: an age and gender matched case-control study. Cancer 1996;77: Asaka M, Kimura T, Kudo M, Terada H, Mitani S, Miyazaki T, Miki K, Graham DY. Relationship of Helicobacter pylori to pepsinogen in an asymptomatic Japanese population. Gastroenterology 1992; 102: Fukuda H, Saito D, Hayashi S, Hisai H, Ono H, Yoshida S, Oguro Y, Noda T, Sato T, Katoh M, Terada M, Sugimura T. Helicobacter pylori infection, serum pepsinogen level and gastric : a case-control study in Japan. Jpn J Cancer Res 1995;86: Blaser MJ, Pérez-Pérez GI, Kleanthous H, Cover TL, Peek RM, Chyou PH, Stemmermann GN, Nomura A. Infection with Helicobacter pylori strains possessing caga is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995;55: Parsonnet J, Friedman GD, Orentreich N, Vogelman H. Risk for gastric in people with CagA positive or CagA negative Helicobacter pylor infection. Gut 1997;40: Rudi J, Kolb C, Maiwald M, Zuna I, von Herbay A, Galle PR, Stremmel W. Serum antibodies against Helicobacter pylori proteins VacA and CagA are associated with increased risk for gastric adenocarcinoma. Dig Dis Sci 1997;42: Queiroz DMM, Mendes EN, Rocha GA, Oliveira AMR, Oliveira CA, Magalhães PP, Moura SB, Cabral MMDA, Nogueira AMMF. cagapositive Helicobacter pylori and risk for developing gastric carcinoma in Brazil. Int J Cancer 1998;78: Torres J, Pérez-Pérez GI, Leal-Herrera Y, Muñoz O. Infection with CagA Helicobacter pylori strains as a possible predictor of risk in the development of gastric adenocarcinoma in Mexico. Int J Cancer 1998;78: Vorobjova T, Nilsson I, Kull K, Maaroos H-I, Covacci A, Wadström T, Uibo R. CagA protein seropositivity in a random sample of adult population and gastric patients in Estonia. Eur J Gastroenterol Hepatol 1998;10: Shimoyama T, Fukuda S, Tanaka M, Mikami T, Munakata A, Crabtree JE. CagA seropositivity associated with development of gastric in a Japanese population. J Clin Pathol 1998;51: Mitchell HM, Hazell SL, Li YY, Hu PJ. Serological response to specific Helicobacter pylori antigens: antibody against CagA antigen is not predictive of gastric in a developing country. Am J Gastroenterol 1996;91: Graham DY, Yamaoka Y. H pylori and caga: relationships with gastric, duodenal ulcer, and reflux esophagitis and its complications. Helicobacter 1998;3: Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven, Forman D. Helicobacter pylori infection and. Br Med Bull 1998;54: Received April 6, Accepted August 9, Address requests for reprints to: Hermann Brenner, M.D., Department of Epidemiology, University of Ulm, Helmholtzstrasse 22, D Ulm, Germany. fax: (49) Supported in part by the German Federal Ministry and Technology grants , , and and by a grant from the Deutsche Herzhilfe. The authors are responsible for the content of this publication.